The issues of modeling when navigating around obstacles of a mobile robot using machine learning methods are considered: Q-learning, SARSA algorithm, deep Q-learning and double deep Q-learning. The developed software includes the Mobile Robotics Simulation Toolbox, Reinforcement Learning Toolbox, and the Gazebo visualization package for environment simulation. The results of the computational experiment show that for a simulated environment with a size of 17 by 17 cells and an obstacle 12 cells long, training using the SARSA algorithm occurs with better performance than for the others.

An algorithm for avoiding obstacles without the use of machine learning is proposed, and it was shown that the speed of avoiding obstacles using this algorithm is higher than the learning speed using deep Q-learning and double deep Q-learning, but lower than using the SARSA and Q-learning algorithms. . For the proposed algorithm, a numerical experiment was carried out using the robot movement simulation environment in Gazebo 11 and it was shown that cubic obstacles are being avoided faster than cylindrical ones.

The problem of determining the features and setting the problem of mathematical modeling of multirotor aircraft (LA) is considered. Differences of their mathematical models from the classical mathematical models of aircraft and single-rotor helicopter type aircraft are considered. The analysis and substantiation of the forces and moments acting on a multirotor aircraft are carried out, taking into account the peculiarities of considering the corresponding coordinate systems necessary for studying the spatial motion of the aircraft. The problem of controlling the trajectory motion of an aircraft is formulated taking into account the rotation of its structure around the center of mass. Based on the consideration of the structural diagram of one of the most common four-rotor aircraft (quadcopter), a scheme for creating control forces and moments acting on the aircraft, under the influence of which the aircraft’s trajectory in space changes, is substantiated. The main mathematical dependences characterizing the kinematics of the motion of a multi-rotor aircraft are given. On the basis of the analysis carried out, a generalized block diagram of the control process for such an aircraft is substantiated and presented. To test the performance and adequacy of the mathematical model, a study was made of the movement of a quadrocopter in a vertical plane between given points in space in accordance with the law of forced control, which ensures the movement of an aircraft in space with maximum speed. The computer simulation of the obtained analytically mathematical dependences showed that this approach is applicable to the construction of mathematical models of the motion of multirotor type aircraft of various design layouts.

The article discusses the features of the practical implementation of the artificial horizon sensor based on the onboard vision system. Proposed, developed on the basis of well-known applications, a variant of the algorithm for the operation of an unmanned aerial vehicle orientation video system. The problems of automatic detection and determination of the position of the horizon line on changing digital images that make up the video stream from the onboard digital camera are shown. The analysis of the factors influencing the accuracy of estimation of carrier orientation angles using the proposed system is carried out. The results of a practical study are presented, characterizing the degree of influence of the considered factors on the total error. A discrete stochastic mathematical model of an unmanned aerial vehicle orientation system based on an onboard vision system has been developed. The possibility of providing an acceptable level of accuracy of the orientation video system due to certain technical and algorithmic solutions is shown. The conclusion is made about the expediency of using this system in autonomous multisensor navigation systems for unmanned aerial vehicles.

With the spread of diabetes mellitus, diabetic retinopathy (DR) is becoming a major public health problem (especially in developing countries). The long-term complications resulting from DR have a significant impact on patients. Early diagnosis and subsequent treatment can reduce the damage to health. Predictive analytics can be based on the analysis of human retinal images using convolutional neural networks. In this paper, the research focuses on the development of an efficient method for DR detection based on the EfficientNet convolutional neural network, self-learning technology and data augmentation operations. As a result of the experiments, a neural network classifier based on convolutional neural networks is developed, recommendations for data augmentation operations are given. Experiments were performed on the public dataset and showed that it is possible to achieve the proportion of correctly classified objects equal to 97.14 % on the test set from the public dataset.

The purpose of the article is to analyze the methods and means of processing cough sounds to detect lung diseases, as well as to describe the developed system for classifying and detecting cough sounds based on a deep neural network. Four types of machine learning and the use of convolutional neural network (CNN) are considered. Hypermarkets of CNN are given. Varieties of machine learning based on the CNN are discussed. The analysis of works on the methodology and means of processing cough sounds based on the CNN with the reduction of the means used and the accuracy of recognition is carried out. Details of machine learning using the environmental sound classification 50 (ESC-50) dataset are discussed. To recognize COVID-19 cough, a classifier was analyzed using CNN as a machine learning model. The proposed CNN system is designed to classify and detect cough sounds based on ESC-50. After selecting a set of sound classification data, four stages are described: extraction of features from audio files, labeling, training, testing. The ESC-50 used for the study was downloaded from the Kaggle website. Python libraries and modules related to deep learning and data science were used to implement the project: NumPy, Librosa, Matplotlib, Hickle, Sci-Kit Learn, Keras. The implemented network used a stochastic gradient algorithm. Several volunteers recorded their voices while coughing using their smartphones and it was assured to record their voices in a public environment to introduce noise to the sounds, in addition to some audio files that were downloaded online. The results showed an average accuracy of 85.37 %, precision of 78.8 % and a recall record of 91.9 %.

The article presents results of our experiments carried out to study the invariance of the digital description of the imageThere in the paper is formulated a mathematical problem of multi-hypothetical detection of subclinical and clinical mastitis in dairy cows by the maximum values of udder temperature measured by digital processing of the udder thermal images. The optimal temperature threshold values corresponding to the Bayesian criterion of the minimum average risk in the above multi-hypothesis detection problem are determined by numerical modelling.

Now one of the most important tasks is development and adaptation of iterative methods for the solution of the superbig rarefied systems of the algebraic equations. The problem of iterative parallel reconstruction of three-dimensional images of industrial facilities leads to such computing tasks. The fact that iterative methods of the solution of computing problems of big dimension are implemented on parallel structures much more effectively, than direct methods of their decision is important.

Physically unclonable functions (PUFs) are basic physical cryptographical primitives, providing to solve tasks such as unclonable identification, digital device authentication and copyright authentication, true random sequence generation, etc. The major features of PUFs are stability, unpredictability and irreproducibility, due to uncontrollable random variations of distinctive features of the raw materials and technological processes used during their manufacturing. Generally, PUF are digital circuits that extract such variations and convert them into a binary format, which applied for further use. Among the variety of PUF types, an Arbiter PUF (APUF) is distinguished, which is a digital circuit with N-bit challenge input and single output for one-bit response generation. The functionality of APUF is based on comparison of transition time of two copies of the test signal along a pair of configurable paths, selected by the challenge value CH from a set of 2N all possible pairs. The result of the comparison is the binary value of the response. The set of all challenge-response pairs is a random, unpredictable and irreproducible in the cases of implementation of cloned PUF circuits both on single and/or on another chips, also using different technologies. This article presents a new approach to the synthesis of the APUF circuits, based on the permutation network elements, which allow to construct the nonlinear structures of pair of paths. This implies the potential complication of building an APUF model to attack its implemented instances. This article presents new schematic solutions for the synthesis of APUF circuits. Also, the main characteristics of the proposed APUF circuits implemented on the Xilinx Zynq-7000 FPGA is analyzed.